architecture-review

Validate architecture markdown files against source code implementations.

Updated May 6, 2026
One-click install
npx skills add https://github.com/dbowm91/codegg --skill architecture-review-dbowm91
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: architecture-review
Source: https://github.com/dbowm91/codegg/tree/main/.opencode/skills/architecture-review
Command: npx skills add https://github.com/dbowm91/codegg --skill architecture-review-dbowm91

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the common issue of documentation drift, where architecture guides become outdated as the codebase evolves, leading to confusion and technical debt.

Core Features & Use Cases

  • Systematic Verification: Automatically cross-references architecture documentation with actual source code to identify discrepancies.
  • Drift Detection: Flags stale line numbers, incorrect counts, missing fields, and phantom types that no longer exist in the implementation.
  • Use Case: Before a major release, run this skill to ensure that the architecture overview and module tables accurately reflect the current state of the codegg codebase.

Quick Start

Initiate the architecture review process by instructing the agent to execute the architecture-review skill across all defined batches.

Frequently Asked Questions about architecture-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify architecture documentation against my source code?

You can verify architecture documentation against source code by systematically cross-referencing markdown files with the project's directory structure and module definitions to identify technical inconsistencies and code-doc drift.

How does documentation drift detection work for codebases?

Documentation drift detection works by parsing architecture markdown files and validating concrete claims against actual source code implementations, automatically flagging stale line numbers, incorrect counts, missing fields, and phantom types.

What's the best way to find stale architecture content before a major release?

The best way to find stale architecture content is to run systematic verification across all defined batches using parallel subagent execution, which produces structured reports highlighting discrepancies between your guides and implementation.

Can I check if my architecture module tables accurately reflect the current codebase?

Yes, you can check if architecture module tables reflect the codebase by executing an architecture review process that validates concrete claims against module definitions and reports technical inconsistencies.

Why does my architecture documentation have phantom types and incorrect counts?

Architecture documentation develops phantom types and incorrect counts due to documentation drift, where guides become outdated as the codebase evolves, leading to confusion and technical debt over time.

Does the architecture review process work with Rust projects?

Yes, the architecture review process supports Rust projects by parsing architecture markdown files and verifying concrete claims against the project's directory structure and module definitions to identify code-doc drift.